INDEX

Explanations

solo debut

The neuron responds to single-word nouns that name a medium, venue, format, or geographic/legal context (e.g. “PC,” “online,” “restaurant,” “London,” “District”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Française

-0.98

itively

-0.90

observers

-0.83

 poule

-0.83

 beaux

-0.83

 hotell

-0.82

[]}

-0.81

сован

-0.81

};

-0.81

 Gefäng

-0.80

POSITIVE LOGITS

 version

0.81

 enfermos

0.81

 despues

0.80

に行きました

0.77

 respald

0.77

 instead

0.77

のですね

0.76

暧

0.76

uak

0.76

Instead

0.75

Activations Density 0.250%